If you’ve ever been to the eye doctors, you’ve undoubtedly been asked, which looks best: A or B. The eye doctor is hoping to gain a better understanding of the intricacies of your eye sight so he can best decipher your vision needs.
What you may not know is that this same sort of test, known as multi-variant or A/B testing (in some circles it’s even referred to as bucket testing), is also used in business.
Simply put, A/B testing is used to test various elements of your digital marketing campaign, giving you insight into your customers’ preferences. Oli Gardner, cofounder of Unbounce.com, the DIY landing page platform, encourages all business to test. “Testing allows you to see which messaging resonates best with your customers, which is going to enable you to position your business in a way that more of your target customers will understand,” he says.
The results from the testing can mean the difference between the customer bouncing from your site and landing the conversion. A/B testing should ideally be run with every marketing campaign in order to create a better optimized sales funnel.
For example, if you’re running an e-mail campaign, you can create two or more variations of your landing page to test. Gardner recommends sending 10 to 30 percent of your list to the test pages. Once you can establish which variant performed better, you can send the remainder of your e-mail list to the winning page.
What Can Be Tested?
Nearly every element of your digital campaign can be tested. But, if you’re not sure where to start, Gardener offers some insight into the key elements which you should focus on initially:
Main headline: This typically contains a succinct rendering of your product, offer or service.
The call to action (CTA): This is typically the text on the button that represents your page’s conversion goal.
Hero shot: Try a variation of your main photo (if you have one), preferably showing your product or service being used in context—video is even better for this—so you should try testing an image and a video.
Form length: For lead capture and other form usage, you will want to minimize the amount of ?elds that visitors are required to complete. However, if you have a particularly strong need for data, try running an A|B|C|D|E test with varying amounts of information gathering. This way you can make an informed decision about what abandonment rate is acceptable when weighed against the extra data produced.
Button design: Use design principles to accentuate the appearance of your CTA (directional cues like arrows, contrast, white-space, size). Above all, try making it stand out more.
Button color: Consider changing your CTA button color to subconsciously encourage engagement: green for go, blue for link color, orange or red for emotional reaction.
Long copy vs. short copy: Often, shorter is better, but for certain products detail is important in the decision making process. Test it and see.
Be careful not to test too many elements at one time. It may be beneficial for future campaigns to know which component your customers liked (or didn’t like) and having multiple elements being tested together will leave you guessing which was the magic bullet.
Blood, Sweat and Tears Often Produces Bias
Why is A/B testing so important? Because your personal investment—the hours of work put into launching a campaign or designing a website or landing page—often establishes a bias of sorts. What you think, and what others within your organization believe to be the best may not be what your customers believe. “If you just stick with conjecture or the opinion of people who are too familiar with your brand, you’re not going to improve your conversions or learn about your customers perception of your brand as quickly,” Gardner says.
Test, Test, Test
Even if you’re not running a promotional campaign, you can test pages of your website in order to maximize viewer engagement. Gardner suggests using analytics software like Google Analytics to see which pages receive the most volume.
Again, do not test all the pages at once. Doing so could cause confusion as to what changes actually increased engagement and conversions.
A/B testing isn’t much different from the blind taste tests between Coke and Pepsi. Although the two are very similar, there are distinct differences that make one preferable over the other to many consumers.
You may never know if your customers prefer Coke or Pepsi (unless, of course, you sell soda) but you may be able to decipher if they respond better to green or red CTA buttons or short or long forms.
Have you ever run an A/B test? Were you able to increase conversions simply by gaining a better understanding of your customers’ preferences?
Angela Stringfellow is a PR and MarComm Consultant and Social Media Strategist offering full-circle marketing solutions to businesses. Angela blogs via Contently.com.
Photo credit: Thinkstock